12ax7 perfect bias point

[Alexrmc92]

First off i would like to introduce myself. my name is Alex and i live in the United States. I have had a fascination with tubes since i was a child and now that i am working towards my degree in electrical engineering i have built many tube guitar amps.

So i am in the process of building my own amp from scratch. I wan't to fix bias the pre amp tube (12AX7) and get a true class A design working.



So i want to make sure that i have all of this right so heres how im looking at it.

The maximum power this tube can dissipate is said to be 1 watt according to the datasheet. My input signal at a maximum will hit 3vpp and i do want distortion after 3vpp. Now since i want the bias to be centered and my input signal at a max will hit 3vpp, i need to bias the grid at -1.5v. Also i need as close to 0ma of current to flow when the grid voltage is at -3v so ill get clipping on the negative side of the input signal.

When the plate voltage is a 205vdc the graph show that 0ma of current will flow were the grid voltage is -3v. To get the current i will require for 1 watt of dissipation i divide 205 / 1 = 4.87ma.

So my load line will go between 4.9ma and 205v. to get the plate resistor value ill do 205 / .0049 and ill get 42kohms.

So the question is this, If i have a 12ax7 with all of these parameters, which as far as i can see is perfectly balanced for a 3vpp signal because it will clip after -3v and after 0v, is it setup properly for operation?

As far as desirable distortion goes in guitar amps, im pretty sure that most of it comes from some type of clipping and i need it to clip equally on the bottom (-3v) and top (0v) of the wave.

[DF96]

You are unlikely to get equal clipping as you are relying on two quite different phenomena: valve cutoff (at -3V) and grid current (somewhere between -1V and 0V). There is no such thing as a perfect bias point, merely the optimum compromise given particular requirements. Bear in mind that the datasheet graphs don't tell you what any particular 12AX7 will do, but merely something like an average or typical response.

If you want carefully defined symmetrical clipping then you need to use a pair of biased diodes. In most cases this is not what you want in a guitar amp, but I will leave the rest to experts.

[revintage]

Why do you want symmetrical clipping? The best "sounding" clipping distortion for guitar usually is when it is unsymmetrical. And why just at 3V?

[costis_n]

Well, if it is for the input stage of the amp. don't have to bias it that cold, almost all designs have them biased at -1.5 V. However, keep in mind that in 12AX7 grid current starts at around -0.5 to -0.9 V.
And do not obsess in minor changes from datasheet to application, it will still work fine. You 'd have to fcuk it up pretty bad to make it not work...

[Structo]

On my amps I like to have a couple clean stages, then a couple more to overdrive them.
So when the overdrive is on, I have four cascading stages before the phase inverter. (LTP)

This is with two 12ax7's.
I use 220K Rp on the first stage and 150K Rp on second, with 3K3 and 2K2 corresponding cathode resistors.
Plate voltage at the tube pins around 200v.'

That seems to give me a very nice clean sound and OD sound.

[costis_n]

Originally Posted by Structo

On my amps I like to have a couple clean stages, then a couple more to overdrive them.
So when the overdrive is on, I have four cascading stages before the phase inverter. (LTP)

This is with two 12ax7's.
I use 220K Rp on the first stage and 150K Rp on second, with 3K3 and 2K2 corresponding cathode resistors.
Plate voltage at the tube pins around 200v.'

That seems to give me a very nice clean sound and OD sound.

Question on tone: Since you run the 2 first stages clean and center biased, it seems it does not really matter if it's a 12AX7 or some other tube. right?

[Alexrmc92]

Originally Posted by revintage

Why do you want symmetrical clipping? The best "sounding" clipping distortion for guitar usually is when it is unsymmetrical. And why just at 3V?

3vpp is typically the accepted value to start clipping, or 1.5vp. Secondly from what I've read / experimented with, asymmetrical clipping is raspy and edgy and sounds more like today's distortion pedals. Symmetrical clipping is more rounded and full and sounds more like 50s-70s tube amps that are being over driven or like tube screamer pedals.

Originally Posted by Structo

On my amps I like to have a couple clean stages, then a couple more to overdrive them.
So when the overdrive is on, I have four cascading stages before the phase inverter. (LTP)

This is with two 12ax7's.
I use 220K Rp on the first stage and 150K Rp on second, with 3K3 and 2K2 corresponding cathode resistors.
Plate voltage at the tube pins around 200v.'

That seems to give me a very nice clean sound and OD sound.

I agree that 4 stages each with a gradual climb in gain instead of two with huge climbs in gain is probably much better as far as linear tube response goes (or as close to linear as possible)

Don't most class AB power tubes clip symmetrically? because if both tubes clip simultaneously then the signal is put together as symmetrically clipping?

Unless i'm wrong, isn't all overdrive or distortion in guitar amps caused by clipping? Maybe not hard clipping into sharp square waves, but more of flattening the peaks and valleys of the wave.


And also aren't the two phenomena i'm looking for called saturation and cutoff? At 0v all of the current from the cathode will go straight towards the plate because grid will no longer be repelling any electrons back. Sure some current will go to the grid but wont most go to the plate?

I know every tube is different and that all of these numbers will be skewed when this is operating in real life. But if my goal is to get distortion when the guitars signal is over 3vpp then do i have it biased right? From what i understand 3vpp is strumming a guitar pretty hard.



This is my first shot at building one from scratch, i understand how they work pretty well. I just want to build my own. I just want to be sure my logic is correct.

[taj]

Hi Alex,

Someone here recently showed me this website for a guy who really studied distortion spectrum. Naturally, his perspective might not match yours or mine (he's after an Eric Clapton sound), but the information regarding distortion technique is really interesting. Read through his guitar amp projects.

Bob Richards Audio Index

..Todd

[Alexrmc92]

Originally Posted by taj

Hi Alex,

Someone here recently showed me this website for a guy who really studied distortion spectrum. Naturally, his perspective might not match yours or mine (he's after an Eric Clapton sound), but the information regarding distortion technique is really interesting. Read through his guitar amp projects.

Bob Richards Audio Index

..Todd

Thank you very much for this. it seems were both in seeking the same type of distortion (eric clapton style) Very interesting ideas in there.

[revintage]

Secondly from what I've read / experimented with, asymmetrical clipping is raspy and edgy and sounds more like today's distortion pedals.

Note asymmetrical working point generates more 2nd harmonic, and that is usually characterised as "round". You might have some other issues that makes it edgy.

By the way, in our production amps we don´t use triodes at all at the input. The input pentode/cascode is actually with our TULP circuit that lets us run them continously variable from triode to pentode characteristics. And with asymmetrical workingpoint, no sight of edgy, harsh sound.